Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/25296 
Year of Publication: 
2008
Series/Report no.: 
SFB 649 Discussion Paper No. 2008,053
Publisher: 
Humboldt University of Berlin, Collaborative Research Center 649 - Economic Risk, Berlin
Abstract: 
We introduce a Nelson-Siegel type interest rate term structure model with the underlying yield factors following autoregressive processes revealing time-varying stochastic volatility. The factor volatilities capture risk inherent to the term struc- ture and are associated with the time-varying uncertainty of the yield curve's level, slope and curvature. Estimating the model based on U.S. government bond yields applying Markov chain Monte Carlo techniques we find that the yield factors and factor volatilities follow highly persistent processes. Using the extracted factors to explain one-year-ahead bond excess returns we observe that the slope and cur- vature yield factors contain the same explanatory power as the return-forecasting factor recently proposed by Cochrane and Piazzesi (2005). Moreover, we identify slope and curvature risk as important additional determinants of future excess returns. Finally, we illustrate that the yield and volatility factors are closely con- nected to variables reflecting macroeconomic activity, inflation, monetary policy and employment growth. It is shown that the extracted yield curve components have long-term prediction power for macroeconomic fundamentals.
Subjects: 
Term structure modelling
yield curve risk
stochastic volatility
factor models
macroeconomic fundamentals
JEL: 
C5
E4
G1
Document Type: 
Working Paper

Files in This Item:
File
Size
737.05 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.